Advanced techniques can be used when there is trend or seasonality, or when other factors (such as price discounts) must be considered.
- Analogous to single regression, but allows us to have multiple predictor variables:
- Y = a + b1*X1 + b2*X2 + b3*X3 …
*Practically speaking, there is a limit to the number of predictor variables you can have without violating some statistical rules.
- In most cases, 2 or 3 predictor variables should be plenty.
In this case, we have 24 months of data.
In addition to an apparent upward trend, we have price discount information and seasonality in the last two months of each year.
Let’s develop a multiple regression forecast model that considers all these factors…
Demand = 9117.08
+ 275.41(Time Period)
+ 2586.31(Seasonal Bump*)
*= 1 if seasonal bump is present; 0 otherwise
The multiple regression model does a decent job modeling past demand. By plugging in the appropriate time period and seasonality value (0 or 1) we can use it to forecast future demands.